Abstract
Identification of mitotic cells as well as estimation of mitotic index are important parameters in understanding the pathology of cancer, predicting response to chemotherapy and overall survival. This is usually performed manually by pathologists and there can be considerable variability in their assessments. The use of deep learning(DL) models can help in addressing this issue. However, most of the state-of-the-art methods are trained for specific cancer types, and often tend to fail when used across multiple tumor types. Hence there is a clear need for a more ‘pan-tumor’ approach to identifying mitotic figures. We propose a generalized DL model for mitosis detection using the MIDOG-2022 Challenge dataset. Using an ensemble of predictions from a transformer-based object detector and a separate classifier, our model makes final predictions. Our approach achieved an F1-score of 0.7569 and stood second in the MIDOG-2022 challenge. The predictions from the object detector alone achieved an F1-score of 0.7510. Our model generalizes well to address the domain shifts caused by variability in image acquisition, protocols and tumor tissue types.
S. Kotte and V.G. Saipradeep—Joint first author.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Dawson, H.: Digital pathology - Rising to the challenge. Front. Med. (Lausanne) 9 888896, (2022)
Jiménez, G., Racoceanu, D.: Classification in computational pathology: application to mitosis analysis in breast cancer grading. Front Bioeng Biotechnol 7(145), (2019)
Veta, M., et al.: Predicting breast tumor proliferation from whole-slide images: the TUPAC16 challenge. Med. Image Anal. 54(145), 111–121 (2019)
Mitos & Atypia 14 contest home page, https://mitos-atypia14.grand-challenge.org/home/. Accessed 4 Sept 2022
Aubreville, M., et al.: Mitosis domain generalization in histopathology images - the MIDOG challenge. Med. Image Anal. 84, 102699 (2023)
Aubreville, M., et al.: Mitosis Domain generalization challenge. In: 2022–25th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2022). Zenodo. https://doi.org/10.5281/zenodo.6362337
Aubreville, M., et al.: MItosis DOmain generalization challenge 2022 (MICCAI MIDOG 2022), Training data set (PNG version) (1.0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.6547151
Tellez, D., et al.: Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology. Med Image Anal. 58, 101544 (2022). Epub 2019 Aug 21. PMID: 31466046. https://doi.org/10.1016/j.media.2019.101544
Carion, N., et al.: End-to-End Object Detection with Transformers. arxiv (2005.12872v3), (2020)
Adam, W.: https://arxiv.org/abs/1711.05101. Accessed 29 Aug 2022
Akiba, T., Sano, S., Yanase, T., Ohta, T., Koyama, M.: Optuna: A next-generation hyperparameter optimization framework. In: Proceedings: 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (2019)
Tan, M., Le, QV.: EfficientNet: rethinking model scaling for convolutional neural networks. In: Proceedings: ICML (2019)
Kingma, D., Ba, J.: Adam: a Method for stochastic optimization. In: 3rd International Conference on Learning Representations (ICLR 2015) Proceedings. ICLR, San Diego, CA, USA (2015)
MIDOG 2022 results. https://midog2022.grand-challenge.org/evaluation/final-test-phase-task-1-without-additional-data/leaderboard/. Accessed 21 Sept 2022
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Kotte, S. et al. (2023). A Deep Learning Based Ensemble Model for Generalized Mitosis Detection in H &E Stained Whole Slide Images. In: Sheng, B., Aubreville, M. (eds) Mitosis Domain Generalization and Diabetic Retinopathy Analysis. MIDOG DRAC 2022 2022. Lecture Notes in Computer Science, vol 13597. Springer, Cham. https://doi.org/10.1007/978-3-031-33658-4_23
Download citation
DOI: https://doi.org/10.1007/978-3-031-33658-4_23
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-33657-7
Online ISBN: 978-3-031-33658-4
eBook Packages: Computer ScienceComputer Science (R0)